26
GRID VGPU FOR VMWARE VSPHERE VERSION 367.106/370.12 RN-07347-001 _v4.3 (GRID) Revision 02 | June 2017 Release Notes

GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Embed Size (px)

Citation preview

Page 1: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

GRID VGPU FOR VMWARE VSPHEREVERSION 367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | June 2017

Release Notes

Page 2: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | ii

TABLE OF CONTENTS

Chapter 1. Release Notes...................................................................................... 1Chapter 2. Validated Platforms................................................................................2

2.1. Supported NVIDIA GPUs and Validated Server Platforms........................................... 22.2. Hypervisor Software Versions........................................................................... 22.3. Guest OS Support......................................................................................... 3

2.3.1. Windows Guest OS Support........................................................................ 32.3.2. Linux Guest OS Support............................................................................ 3

Chapter 3. Known Product Limitations......................................................................53.1. vGPU profiles with 512 Mbytes or less of frame buffer support only 1 virtual display head

on Windows 10.............................................................................................. 53.2. NVENC requires at least 1 Gbyte of frame buffer.................................................. 63.3. VM running older NVIDIA vGPU drivers fails to initialize vGPU when booted....................63.4. Virtual GPU fails to start if ECC is enabled..........................................................73.5. Single vGPU benchmark scores are lower than passthrough GPU.................................83.6. GRID K1 and GRID K2 cards do not support monitoring of vGPU engine usage..................93.7. VMs configured with large memory fail to initialize vGPU when booted....................... 10

Chapter 4. Resolved Issues................................................................................... 12Chapter 5. Known Issues......................................................................................13

5.1. Memory exhaustion can occur with vGPU profiles that have 512 Mbytes or less of framebuffer........................................................................................................13

5.2. vGPU VM fails to boot in ESXi 6.5 if the graphics type is Shared................................145.3. ESXi 6.5 web client shows high memory usage even when VMs are idle....................... 155.4. GRID Virtual GPU Manager must not be on a host in a VMware DRS cluster................... 165.5. GNOME Display Manager (GDM) fails to start on Red Hat Enterprise Linux 7.2 and CentOS

7.0............................................................................................................175.6. NVIDIA Control Panel fails to start and reports that “you are not currently using a display

that is attached to an Nvidia GPU”....................................................................175.7. VM configured with more than one vGPU fails to initialize vGPU when booted............... 185.8. A VM configured with both a vGPU and a passthrough GPU fails to start the passthrough

GPU.......................................................................................................... 195.9. vGPU allocation policy fails when multiple VMs are started simultaneously...................195.10. Before Horizon agent is installed inside a VM, the Start menu’s sleep option is available.. 205.11. vGPU-enabled VMs fail to start, nvidia-smi fails when VMs are configured with too high

a proportion of the server’s memory.................................................................. 205.12. On reset or restart VMs fail to start with the error VMIOP: no graphics device is available

for vGPU….................................................................................................. 215.13. nvidia-smi shows high GPU utilization for vGPU VMs with active Horizon sessions.......... 225.14. Multiple WebGL tabs in Microsoft Internet Explorer may trigger TDR on Windows VMs.... 22

Page 3: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 1

Chapter 1.RELEASE NOTES

These Release Notes summarize current status, information on validated platforms,and known issues with NVIDIA GRID™ vGPU™ software and hardware on VMwarevSphere.

This release includes the following software:

‣ NVIDIA GRID Virtual GPU Manager version 367.106 for the VMware vSpherereleases listed in Hypervisor Software Versions

‣ NVIDIA Windows drivers for vGPU version 370.12‣ NVIDIA Linux drivers for vGPU version 367.106

Caution

The GRID vGPU Manager and Windows guest VM drivers must be installed together.Older VM drivers will not function correctly with this release of GRID vGPU Manager.Similarly, older GRID vGPU Managers will not function correctly with this release ofWindows guest drivers. See VM running older NVIDIA vGPU drivers fails to initializevGPU when booted.

Updates in this release:

‣ Miscellaneous bug fixes

Page 4: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 2

Chapter 2.VALIDATED PLATFORMS

This release of virtual GPU provides support for several NVIDIA GPUs on validatedserver hardware platforms, VMware vSphere hypervisor software versions, and guestoperating systems.

2.1. Supported NVIDIA GPUs and Validated ServerPlatformsThis release of virtual GPU provides support for the following NVIDIA GPUs onVMware vSphere, running on validated server hardware platforms:

‣ GRID K1‣ GRID K2‣ Tesla M6‣ Tesla M10‣ Tesla M60

For a list of validated server platforms, refer to NVIDIA GRID Certified Servers.

Tesla M60 and M6 GPUs support compute mode and graphics mode. GRID vGPUrequires GPUs that support both modes to operate in graphics mode.

Recent Tesla M60 GPUs and M6 GPUs are supplied in graphics mode. However, yourGPU might be in compute mode if it is an older Tesla M60 GPU or M6 GPU, or if itsmode has previously been changed.

To configure the mode of Tesla M60 and M6 GPUs, use the gpumodeswitch toolprovided with GRID software releases.

2.2. Hypervisor Software VersionsThis release has been tested with the following hypervisor software versions:

Page 5: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Validated Platforms

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 3

Software Version Tested

VMware vSphere Hypervisor (ESXi) 6.0 RTM build 2494585

6.0 update 1

6.0 update 2

6.5

VMware Horizon 6.2.1 RTM build 3268071

7.0.2 build 4368292

7.1.0 RTM build 5170901

VMware vCenter Server 6.0 RTM build 2562643

6.5.0 RTM build 4602587

2.3. Guest OS SupportGRID vGPU supports several Windows releases and Linux distributions as a guest OS.

Use only a guest OS release that is listed as supported by GRID vGPU with yourvirtualization software. To be listed as supported, a guest OS release must besupported not only by GRID vGPU, but also by your virtualization software. NVIDIAcannot support guest OS releases that your virtualization software does not support.

2.3.1. Windows Guest OS SupportGRID vGPU supports the following Windows releases as a guest OS on VMwarevSphere:

‣ Windows 7 (32/64-bit)‣ Windows 8 (32/64-bit)‣ Windows 8.1 (32/64-bit)‣ Windows 10 (32/64-bit)‣ Windows Server 2008 R2‣ Windows Server 2012 R2‣ Windows Server 2016

2.3.2. Linux Guest OS SupportGRID vGPU supports the following Linux distributions as a guest OS only on supportedTesla GPUs on VMware vSphere:

‣ Red Hat Enterprise Linux 7.0-7.3 and later compatible 7.x versions‣ CentOS 7.0-7.3 and later compatible 7.x versions‣ Red Hat Enterprise Linux 6.6 and later compatible 6.x versions‣ CentOS 6.6 and later compatible 6.x versions‣ Ubuntu 16.04 LTS

Page 6: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Validated Platforms

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 4

‣ Ubuntu 14.04 LTS‣ Ubuntu 12.04 LTS

GRID K1 and GRID K2 do not support vGPU on a Linux guest OS.

Page 7: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 5

Chapter 3.KNOWN PRODUCT LIMITATIONS

Known product limitations for this release of NVIDIA GRID are described in thefollowing sections.

3.1. vGPU profiles with 512 Mbytes or less offrame buffer support only 1 virtual display headon Windows 10

Description

To reduce the possibility of memory exhaustion, vGPU profiles with 512 Mbytes or lessof frame buffer support only 1 virtual display head on a Windows 10 guest OS.

The following vGPU profiles have 512 Mbytes or less of frame buffer:

‣ Tesla M6-0B, M6-0Q‣ Tesla M10-0B, M10-0Q‣ Tesla M60-0B, M60-0Q‣ GRID K100, K120Q‣ GRID K200, K220Q

Workaround

Use a profile that supports more than 1 virtual display head and has at least 1 Gbyte offrame buffer.

Page 8: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Product Limitations

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 6

3.2. NVENC requires at least 1 Gbyte of framebuffer

Description

Using the frame buffer for the NVIDIA hardware-based H.264/HEVC video encoder(NVENC) may cause memory exhaustion with vGPU profiles that have 512 Mbytesor less of frame buffer. To reduce the possibility of memory exhaustion, NVENC isdisabled on profiles that have 512 Mbytes or less of frame buffer. Application GPUacceleration remains fully supported and available for all profiles, including profileswith 512 MBytes or less of frame buffer. NVENC support from both Citrix and VMwareis a recent feature and, if you are using an older version, you should experience nochange in functionality.

The following vGPU profiles have 512 Mbytes or less of frame buffer:

‣ Tesla M6-0B, M6-0Q‣ Tesla M10-0B, M10-0Q‣ Tesla M60-0B, M60-0Q‣ GRID K100, K120Q‣ GRID K200, K220Q

Workaround

If you require NVENC to be enabled, use a profile that has at least 1 Gbyte of framebuffer.

3.3. VM running older NVIDIA vGPU drivers fails toinitialize vGPU when booted

Description

A VM running older NVIDIA drivers, such as those from a previous vGPU release, willfail to initialize vGPU when booted on a VMware vSphere platform running the currentrelease of GRID Virtual GPU Manager.

In this scenario, the VM boots in standard VGA mode with reduced resolution and colordepth. The NVIDIA GRID GPU is present in Windows Device Manager but displays awarning sign, and the following device status:

Windows has stopped this device because it has reported problems. (Code 43)

Page 9: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Product Limitations

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 7

Depending on the versions of drivers in use, the VMware vSphere VM’s log file reportsone of the following errors:

‣ A version mismatch between guest and host drivers:

vthread-10| E105: vmiop_log: Guest VGX version(2.0) and Host VGX version(2.1) do not match

‣ A signature mismatch:

vthread-10| E105: vmiop_log: VGPU message signature mismatch.

Resolution

Install the latest NVIDIA vGPU release drivers in the VM.

3.4. Virtual GPU fails to start if ECC is enabled

Description

GRID K2, Tesla M60, and Tesla M6 support error correcting code (ECC) for improveddata integrity. If ECC is enabled, virtual GPU fails to start. The following error is loggedin the VMware vSphere VM’s log file:

vthread10|E105: Initialization: VGX not supported with ECC Enabled.

Virtual GPU is not currently supported with ECC active. GRID K2 cards and TeslaM60, M6 cards in graphics mode ship with ECC disabled by default, but ECC maysubsequently be enabled using nvidia-smi.

Resolution

Ensure that ECC is disabled on all GPUs.

1. Use nvidia-smi to list the status of all GPUs, and check for ECC noted as enabledon GPUs.

2. Change the ECC status to off on each GPU for which ECC is enabled by executingthe following command:

nvidia-smi -i id -e 0

id is the index of the GPU as reported by nvidia-smi.

Page 10: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Product Limitations

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 8

3.5. Single vGPU benchmark scores are lower thanpassthrough GPU

Description

A single vGPU configured on a physical GPU produces lower benchmark scores thanthe physical GPU run in passthrough mode.

Aside from performance differences that may be attributed to a vGPU’s smallerframebuffer size, vGPU incorporates a performance balancing feature known as FrameRate Limiter (FRL), which is enabled on all vGPUs. FRL is used to ensure balancedperformance across multiple vGPUs that are resident on the same physical GPU. TheFRL setting is designed to give good interactive remote graphics experience but mayreduce scores in benchmarks that depend on measuring frame rendering rates, ascompared to the same benchmarks running on a passthrough GPU.

Resolution

FRL is controlled by an internal vGPU setting. NVIDIA does not validatevGPU with FRL disabled, but for validation of benchmark performance,FRL can be temporarily disabled by adding the configuration parameterpciPassthru0.cfg.frame_rate_limiter in the VM’s advanced configurationoptions.

This setting can only be changed when the VM is powered off.

1. Select Edit Settings. 2. In Edit Settings window, select the VM Options tab. 3. From the Advanced drop-down list, select Edit Configuration. 4. In the Configuration Parameters dialog box, click Add Row. 5. In the Name field, type the parameter name

pciPassthru0.cfg.frame_rate_limiter, in the Value field type 0, and clickOK.

Page 11: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Product Limitations

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 9

With this setting in place, the VM’s vGPU will run without any framerate limit. The FRL can be reverted back to its default setting by settingpciPassthru0.cfg.frame_rate_limiter to 1 or by removing the parameter fromthe advanced settings.

3.6. GRID K1 and GRID K2 cards do not supportmonitoring of vGPU engine usage

Description

GRID K1 and GRID K2 cards do not support monitoring of vGPU engine usage. Alltools and APIs for any vGPU running on GRID K1 or GRID K2 cards report 0 for thefollowing usage statistics:

‣ 3D/Compute‣ Memory controller bandwidth‣ Video encoder‣ Video decoder

Page 12: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Product Limitations

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 10

3.7. VMs configured with large memory fail toinitialize vGPU when booted

Description

When starting multiple VMs configured with large amounts of RAM (typically morethan 32GB per VM), a VM may fail to initialize vGPU. In this scenario, the VM boots inVMware SVGA mode and doesn’t load the NVIDIA driver. The NVIDIA GRID GPU ispresent in Windows Device Manager but displays a warning sign, and the followingdevice status:

Windows has stopped this device because it has reported problems. (Code 43)

The VMware vSphere VM’s log file contains these error messages:

vthread10|E105: NVOS status 0x29vthread10|E105: Assertion Failed at 0x7620fd4b:179vthread10|E105: 8 frames returned by backtrace ...vthread10|E105: VGPU message 12 failed, result code: 0x29...vthread10|E105: NVOS status 0x8vthread10|E105: Assertion Failed at 0x7620c8df:280vthread10|E105: 8 frames returned by backtrace...vthread10|E105: VGPU message 26 failed, result code: 0x8

Resolution

vGPU reserves a portion of the VM’s framebuffer for use in GPU mapping of VM systemmemory. The reservation is sufficient to support up to 32GB of system memory, andmay be increased to accommodate up to 64GB by adding the configuration parameterpciPassthru0.cfg.enable_large_sys_mem in the VM’s advanced configurationoptions

This setting can only be changed when the VM is powered off.

1. Select Edit Settings. 2. In Edit Settings window, select the VM Options tab. 3. From the Advanced drop-down list, select Edit Configuration. 4. In the Configuration Parameters dialog box, click Add Row. 5. In the Name field, type the parameter name

pciPassthru0.cfg.enable_large_sys_mem, in the Value field type 1, andclick OK.

With this setting in place, less GPU framebuffer is available to applications runningin the VM. To accommodate system memory larger than 64GB, the reservation can

Page 13: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Product Limitations

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 11

be further increased by adding pciPassthru0.cfg.extra_fb_reservationin the VM’s advanced configuration options, and setting its value to thedesired reservation size in megabytes. The default value of 64M is sufficientto support 64 GB of RAM. We recommend adding 2 M of reservation for eachadditional 1 GB of system memory. For example, to support 96 GB of RAM, setpciPassthru0.cfg.extra_fb_reservation to 128.

The reservation can be reverted back to its default setting by settingpciPassthru0.cfg.enable_large_sys_mem to 0, or by removing the parameterfrom the advanced settings.

Page 14: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 12

Chapter 4.RESOLVED ISSUES

Bug ID Summary and Description

1816290 The VMware VIB installer incorrectly reports that reboot is not required afterinstalling the vGPU Manager VIB

After installing the NVIDIA Virtual GPU Manager VIB for vSphere on the ESXi host,the esxcli command to install the VIB incorrectly reports Reboot Required:false in the installation result message.

Page 15: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 13

Chapter 5.KNOWN ISSUES

5.1. Memory exhaustion can occur with vGPUprofiles that have 512 Mbytes or less of framebuffer

Description

Memory exhaustion can occur with vGPU profiles that have 512 Mbytes or less of framebuffer.

This issue typically occurs in the following situations:

‣ Full screen 1080p video content is playing in a browser. In this situation, the sessionhangs and session reconnection fails.

‣ Multiple display heads are used with Citrix XenDesktop or VMware Horizon on aWindows 10 guest VM.

‣ Higher resolution monitors are used.‣ Applications that are frame-buffer intensive are used.‣ NVENC is in use.

To reduce the possibility of memory exhaustion, NVENC is disabled on profiles thathave 512 Mbytes or less of frame buffer.

When memory exhaustion occurs, the NVIDIA host driver reports Xid error 31 andXid error 43 in the VMware vSphere log file vmware.log in the guest VM’s storagedirectory.

The following vGPU profiles have 512 Mbytes or less of frame buffer:

‣ Tesla M6-0B, M6-0Q‣ Tesla M10-0B, M10-0Q

Page 16: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 14

‣ Tesla M60-0B, M60-0Q‣ GRID K100, K120Q‣ GRID K200, K220Q

The root cause is a known issue associated with changes to the way that recent Microsoftoperating systems handle and allow access to overprovisioning messages and errors. Ifyour systems are provisioned with enough frame buffer to support your use cases, youshould not encounter these issues.

Workaround

‣ Use an appropriately sized vGPU to ensure that the frame buffer supplied to a VMthrough the vGPU is adequate for your workloads.

‣ Monitor your frame buffer usage.‣ If you are using Windows 10, consider these workarounds and solutions:

‣ Use a profile that has 1 Gbyte of frame buffer.‣ Optimize your Windows 10 resource usage.

To obtain information about best practices for improved user experience usingWindows 10 in virtual environments, complete the NVIDIA GRID vGPU ProfileSizing Guide for Windows 10 download request form.

Additionally, you can use the VMware OS Optimization Tool to make and applyoptimization recommendations for Windows 10 and other operating systems.

Status

Open

Ref. #

‣ 200130864‣ 1803861

5.2. vGPU VM fails to boot in ESXi 6.5 if thegraphics type is Shared

Description

On VMware vSphere Hypervisor (ESXi) 6.5, after vGPU is configured, VMs to which avGPU is assigned may fail to start and the following error message may be displayed:

The amount of graphics resource available in the parent resource pool is insufficient for the operation.

Page 17: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 15

The vGPU Manager VIB provides vSGA and vGPU functionality in a single VIB.After this VIB is installed, the default graphics type is Shared, which provides vSGAfunctionality. To enable vGPU support for VMs in VMware vSphere 6.5, you mustchange the default graphics type to Shared Direct. If you do not change the defaultgraphics type you will encounter this issue.

Version

VMware vSphere Hypervisor (ESXi) 6.5

Workaround

Change the default graphics type to Shared Direct as explained in GRID Virtual GPUUser Guide.

Status

Open

Ref. #

200256224

5.3. ESXi 6.5 web client shows high memory usageeven when VMs are idle

Description

On VMware vSphere Hypervisor (ESXi) 6.5, the web client shows a memory usagealarm with critical severity for VMs to which a vGPU is attached even when the VMs areidle. When memory usage is monitored from inside the VM, no memory usage alarm isshown. The web client does not show a memory usage alarm for the same VMs withoutan attached vGPU.

Version

VMware vSphere Hypervisor (ESXi) 6.5

Workaround

Avoid using the VMware vSphere Hypervisor (ESXi) 6.5 web client to monitor memoryusage for VMs to which a vGPU is attached.

Status

Not an NVIDIA bug

Page 18: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 16

Ref. #

200191065

5.4. GRID Virtual GPU Manager must not be on ahost in a VMware DRS cluster

Description

The ESXi host on which the NVIDIA Virtual GPU Manager for vSphere is installedmust not be a member of a VMware Distributed Resource Scheduler (DRS) cluster. Theinstaller for the NVIDIA driver for GRID Virtual GPU cannot locate the GRID GPU cardon a host in a VMware DRS Cluster. Any attempt to install the driver on a VM on a hostin a DRS cluster fails with the following error:

NVIDIA Installer cannot continueThis graphics driver could not find compatible graphics hardware.

Version

Workaround

Move GRID Virtual GPU Manager to a host outside the DRS cluster.

1. Remove GRID Virtual GPU Manager from the host in the DRS cluster. 2. Create a cluster of VMware ESXi hosts outside the DRS domain. 3. Install the GRID Virtual GPU Manager on an ESXi host in the cluster that you

created in the previous step. 4. Create a vSphere VM for use with GRID Virtual GPU. 5. Configure the vSphere VM with GRID Virtual GPU. 6. Boot the vSphere VM and install the NVIDIA driver for GRID Virtual GPU.

For instructions for performing these tasks, refer to GRID Virtual GPU User Guide.

Status

Open

Ref. #

1933449

Page 19: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 17

5.5. GNOME Display Manager (GDM) fails to starton Red Hat Enterprise Linux 7.2 and CentOS 7.0

Description

GDM fails to start on Red Hat Enterprise Linux 7.2 and CentOS 7.0 with the followingerror:

Oh no! Something has gone wrong!

Workaround

Permanently enable permissive mode for Security Enhanced Linux (SELinux).

1. As root, edit the /etc/selinux/config file to set SELINUX to permissive.SELINUX=permissive

2. Reboot the system.

~]# reboot

For more information, see Permissive Mode in Red Hat Enterprise Linux 7 SELinux User'sand Administrator's Guide.

Status

Not an NVIDIA bug

Ref. #

200167868

5.6. NVIDIA Control Panel fails to start and reportsthat “you are not currently using a display that isattached to an Nvidia GPU”

Description

When you launch NVIDIA Control Panel on a VM configured with vGPU, it fails to startand reports that you are not using a display attached to an NVIDIA GPU. This happensbecause Windows is using VMware’s SVGA device instead of NVIDIA vGPU.

Fix

Make NVIDIA vGPU the primary display adapter.

Page 20: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 18

Use Windows screen resolution control panel to make the second display, identified as“2” and corresponding to NVIDIA vGPU, to be the active display and select the Showdesktop only on 2 option. Click Apply to accept the configuration.

You may need to click on the Detect button for Windows to recognize the displayconnected to NVIDIA vGPU.

If the VMware Horizon/View agent is installed in the VM, the NVIDIA GPU isautomatically selected in preference to the SVGA device.

Status

Open

Ref. #

5.7. VM configured with more than one vGPU failsto initialize vGPU when booted

Description

Using the current VMware vCenter user interface, it is possible to configure a VM withmore than one vGPU device. When booted, the VM boots in VMware SVGA mode anddoesn’t load the NVIDIA driver. The additional vGPU devices are present in WindowsDevice Manager but display a warning sign, and the following device status:

Windows has stopped this device because it has reported problems. (Code 43)

Workaround

GRID vGPU currently supports a single virtual GPU device per VM. Remove anyadditional vGPUs from the VM configuration before booting the VM.

Status

Open

Page 21: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 19

Ref. #

5.8. A VM configured with both a vGPU and apassthrough GPU fails to start the passthroughGPU

Description

Using the current VMware vCenter user interface, it is possible to configure a VM witha vGPU device and a passthrough (direct path) GPU device. This is not a currentlysupported configuration for vGPU. The passthrough GPU appears in Windows DeviceManager with a warning sign, and the following device status:

Windows has stopped this device because it has reported problems. (Code 43)

Workaround

Do not assign vGPU and passthrough GPUs to a VM simultaneously.

Status

Open

Ref. #

1735002

5.9. vGPU allocation policy fails when multipleVMs are started simultaneously

Description

If multiple VMs are started simultaneously, vSphere may not adhere to the placementpolicy currently in effect. For example, if the default placement policy (breadth-first)is in effect, and 4 physical GPUs are available with no resident vGPUs, then starting 4VMs simultaneously should result in one vGPU on each GPU. In practice, more than onevGPU may end up resident on a GPU.

Workaround

Start VMs individually.

Page 22: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 20

Status

Not an NVIDIA bug

Ref. #

200042690

5.10. Before Horizon agent is installed inside aVM, the Start menu’s sleep option is available

Description

When a VM is configured with a vGPU, the Sleep option remains available in theWindows Start menu. Sleep is not supported on vGPU and attempts to use it will leadto undefined behavior.

Workaround

Do not use Sleep with vGPU.

Installing the VMware Horizon agent will disable the Sleep option.

Status

Closed

Ref. #

200043405

5.11. vGPU-enabled VMs fail to start, nvidia-smi fails when VMs are configured with too high aproportion of the server’s memory.

Description

If vGPU-enabled VMs are assigned too high a proportion of the server’s total memory,the following errors occur:

‣ One or more of the VMs may fail to start with the following error:

The available Memory resources in the parent resource pool are insufficient for the operation

‣ When run in the host shell, the nvidia-smi utility returns this error:

Page 23: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 21

-sh: can't fork

For example, on a server configured with 256G of memory, these errors may occur ifvGPU-enabled VMs are assigned more than 243G of memory.

Workaround

Reduce the total amount of system memory assigned to the VMs.

Status

Closed

Ref. #

200060499

5.12. On reset or restart VMs fail to start withthe error VMIOP: no graphics device isavailable for vGPU…

Description

On a system running a maximal configuration, that is, with the maximum number ofvGPU VMs the server can support, some VMs might fail to start post a reset or restartoperation.

Fix

Upgrade to ESXi 6.0 Update 1.

Status

Closed

Ref. #

200097546

Page 24: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 22

5.13. nvidia-smi shows high GPU utilization forvGPU VMs with active Horizon sessions

Description

vGPU VMs with an active Horizon connection utilize a high percentage of the GPU onthe ESXi host. The GPU utilization remains high for the duration of the Horizon sessioneven if there are no active applications running on the VM.

Workaround

None

Status

Open

Partially resolved for Horizon 7.0.1:

‣ For Blast connections, GPU utilization is no longer high.‣ For PCoIP connections, utilization remains high.

Ref. #

1735009

5.14. Multiple WebGL tabs in Microsoft InternetExplorer may trigger TDR on Windows VMs

Description

Running intensive WebGL applications in multiple IE tabs may trigger a TDR onWindows VMs.

Workaround

Disable hardware acceleration in IE.

To enable software rendering in IE, refer to the Microsoft knowledge base article How toenable or disable software rendering in Internet Explorer.

Status

Open

Page 25: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Known Issues

www.nvidia.comGRID vGPU for VMware vSphere Version367.106/370.12

RN-07347-001 _v4.3 (GRID) Revision 02 | 23

Ref. #

200148377

Page 26: GRID vGPU for VMware vSphere Version 367.106/370 · PDF filevthread10|E105: Initialization: VGX not supported with ECC Enabled. Virtual GPU is not currently supported with ECC active

Notice

ALL NVIDIA DESIGN SPECIFICATIONS, REFERENCE BOARDS, FILES, DRAWINGS,DIAGNOSTICS, LISTS, AND OTHER DOCUMENTS (TOGETHER AND SEPARATELY,"MATERIALS") ARE BEING PROVIDED "AS IS." NVIDIA MAKES NO WARRANTIES,EXPRESSED, IMPLIED, STATUTORY, OR OTHERWISE WITH RESPECT TO THEMATERIALS, AND EXPRESSLY DISCLAIMS ALL IMPLIED WARRANTIES OFNONINFRINGEMENT, MERCHANTABILITY, AND FITNESS FOR A PARTICULARPURPOSE.

Information furnished is believed to be accurate and reliable. However, NVIDIACorporation assumes no responsibility for the consequences of use of suchinformation or for any infringement of patents or other rights of third partiesthat may result from its use. No license is granted by implication of otherwiseunder any patent rights of NVIDIA Corporation. Specifications mentioned in thispublication are subject to change without notice. This publication supersedes andreplaces all other information previously supplied. NVIDIA Corporation productsare not authorized as critical components in life support devices or systemswithout express written approval of NVIDIA Corporation.

HDMI

HDMI, the HDMI logo, and High-Definition Multimedia Interface are trademarks orregistered trademarks of HDMI Licensing LLC.

OpenCL

OpenCL is a trademark of Apple Inc. used under license to the Khronos Group Inc.

Trademarks

NVIDIA and the NVIDIA logo are trademarks or registered trademarks of NVIDIACorporation in the U.S. and other countries. Other company and product namesmay be trademarks of the respective companies with which they are associated.

Copyright

© 2013-2017 NVIDIA Corporation. All rights reserved.

www.nvidia.com